“…In [47], the authors propose an ensemble of data-driven prognostic algorithms for robust prediction of RUL by the weighted sum of the outputs of five different methods: a Similarity-Based Interpolation (SBI) approach with Relevance Vector Machine (RVM) as the regression technique, SBI with Support Vector Machine (SVM), SBI with the least-square exponential fitting, a Bayesian linear regression with the least-square quadratic fitting, and a Recurrent Neural Network (RNN) approach. Specifically on fault prognosis in power transformers, it is possible to find some few recent studies in the literature, as the ones described in [35,36].…”